BERT 기반 감성분석을 이용한 추천시스템

박호연(Ho-yeon Park),김경재(Kyoung-jae Kim) 한국지능정보시스템학회 2021 지능정보연구 Vol.27 No.2

If it is difficult for us to make decisions, we ask for advice from friends or people around us. When we decide to buy products online, we read anonymous reviews and buy them. With the advent of the Data-driven era, IT technologys development is spilling out many data from individuals to objects. Companies or individuals have accumulated, processed, and analyzed such a large amount of data that they can now make decisions or execute directly using data that used to depend on experts. Nowadays, the recommender system plays a vital role in determining the users preferences to purchase goods and uses a recommender system to induce clicks on web services (Facebook, Amazon, Netflix, Youtube). For example, Youtubes recommender system, which is used by 1 billion people worldwide every month, includes videos that users like, like and videos they watched. Recommended system research is deeply linked to practical business. Therefore, many researchers are interested in building better solutions. Recommender systems use the information obtained from their users to generate recommendations because the development of the provided recommender systems requires information on items that are likely to be preferred by the user. We began to trust patterns and rules derived from data rather than empirical intuition through the recommender systems. The capacity and development of data have led machine learning to develop deep learning. However, such recommender systems are not all solutions. Proceeding with the recommender systems, there should be no scarcity in all data and a sufficient amount. Also, it requires detailed information about the individual. The recommender systems work correctly when these conditions operate. The recommender systems become a complex problem for both consumers and sellers when the interaction log is insufficient. Because the sellers perspective needs to make recommendations at a personal level to the consumer and receive appropriate recommendations with reliable data from the consumers perspective. In this paper, to improve the accuracy problem for appropriate recommendation to consumers, the recommender systems are proposed in combination with context-based deep learning. This research is to combine user-based data to create hybrid Recommender Systems. The hybrid approach developed is not a collaborative type of Recommender Systems, but a collaborative extension that integrates user data with deep learning. Customer review data were used for the data set. Consumers buy products in online shopping malls and then evaluate product reviews. Rating reviews are based on reviews from buyers who have already purchased, giving users confidence before purchasing the product. However, the recommendation system mainly uses scores or ratings rather than reviews to suggest items purchased by many users. In fact, consumer reviews include product opinions and user sentiment that will be spent on evaluation. By incorporating these parts into the study, this paper aims to improve the recommendation system. This study is an algorithm used when individuals have difficulty in selecting an item. Consumer reviews and record patterns made it possible to rely on recommendations appropriately. The algorithm implements a recommendation system through collaborative filtering. This studys predictive accuracy is measured by Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Netflix is strategically using the referral system in its programs through competitions that reduce RMSE every year, making fair use of predictive accuracy. Research on hybrid recommender systems combining the NLP approach for personalization recommender systems, deep learning base, etc. has been increasing. Among NLP studies, sentiment analysis began to take shape in the mid-2000s as user review data increased. Sentiment analysis is a text classification task based on machine learning. The machine learning-based sentiment analysis has a disadvan

사용자 선호도 변화에 따른 추천시스템의 다양성 적용

나혜연(Hyeyeon Na),남기환(Kihwan Nam) 한국지능정보시스템학회 2020 지능정보연구 Vol.26 No.4

Recommender Systems have been huge influence users and business more and more. Recently the importance of E-commerce has been reached rapid growth greatly in world-wide COVID-19 pandemic. Recommender system is the center of E-commerce lively. Top ranked E-commerce managers mentioned that recommender systems have a major influence on customer’s purchase such as about 50% of Netflix, Amazon sales from their recommender systems. Most algorithms have been focused on improving accuracy of recommender system regardless of novelty, diversity, serendipity etc. Recommender systems with only high accuracy cannot satisfy business long-term profit because of generating sales polarization. In addition, customers do not experience enjoyment of shopping from only focusing accuracy recommender system because customer’s preference is changed constantly. Therefore, recommender systems with various values need to be developed for user’s high satisfaction. Reranking is the most useful methodology to realize diversity of recommender system. In this paper, diversity of recommender system is represented through constructing high similarity with users who have different preference using each user’s purchased item’s category algorithm. It is distinguished from past research approach which is changing the algorithm of recommender system without user’s diversity preference level. We tried to discover user’s diversity preference level and observed the results how the effect was different according to user’s diversity preference level. In addition, graph-based recommender system was used to show diversity through user’s network, not collaborative filtering. In this paper, Amazon Grocery and Gourmet Food data was used because the low-involvement product, such as habitual product, foods, low-priced goods etc., had high probability to show customer’s diversity. First, a bipartite graph with users and items simultaneously is constructed to make graph-based recommender system. However, each users and items unipartite graph also need to be established to show diversity of recommender system. The weight of each unipartite graph has played crucial role changing Jaccard Distance of item’s category. We can observe two important results from the user’s unipartite network. First, the user’s diversity preference level is observed from the network and second, dissimilar users can be discovered in the user’s network. Through the research process, diversity of recommender system is presented highly with small accuracy loss and optimalization for higher accuracy is possible controlling diversity ratio. This paper has three important theoretical points. First, this research expands recommender system research for user’s satisfaction with various values. Second, the graph-based recommender system is developed newly. Third, the evaluation indicator of diversity is made for diversity. In addition, recommender systems are useful for corporate profit practically and this paper has contribution on business closely. Above all, business long-term profit can be improved using recommender system with diversity and the recommender system can provide right service according to user’s diversity level. Lastly, the corporate selling low-involvement products have great effect based on the results.

A Topic Modeling-based Recommender System Consider ing Changes in User Preferences

So Young Kang(강소영),Jae Kyeong Kim(김재경),Il Young Choi(최일영),Chang Dong Kang(강창동) 한국지능정보시스템학회 2020 지능정보연구 Vol.26 No.2

Recommender systems help users make the best choice among various options. Especially, recommender systems play important roles in internet sites as digital information is generated innumerable every second. Many studies on recommender systems have focused on an accurate recommendation. However, there are some problems to overcome in order for the recommendation system to be commercially successful. First, there is a lack of transparency in the recommender system. That is, users cannot know why products are recommended. Second, the recommender system cannot immediately reflect changes in user preferences. That is, although the preference of the users product changes over time, the recommender system must rebuild the model to reflect the users preference. Therefore, in this study, we proposed a recommendation methodology using topic modeling and sequential association rule mining to solve these problems from review data. Product reviews provide useful information for recommendations because product reviews include not only rating of the product but also various contents such as user experiences and emotional state. So, reviews imply user preference for the product. So, topic modeling is useful for explaining why items are recommended to users. In addition, sequential association rule mining is useful for identifying changes in user preferences. The proposed methodology is largely divided into two phases. The first phase is to create user profile based on topic modeling. After extracting topics from user reviews on products, user profile on topics is created. The second phase is to recommend products using sequential rules that appear in buying behaviors of users as time passes. The buying behaviors are derived from a change in the topic of each user. A collaborative filtering-based recommendation system was developed as a benchmark system, and we compared the performance of the proposed methodology with that of the collaborative filtering-based recommendation system using Amazons review dataset. As evaluation metrics, accuracy, recall, precision, and F1 were used. For topic modeling, collapsed Gibbs sampling was conducted. And we extracted 15 topics. Looking at the main topics, topic 1, top 3, topic 4, topic 7, topic 9, topic 13, topic 14 are related to “comedy shows”, “high-teen drama series”, “crime investigation drama”, “horror theme”, “British drama”, “medical drama”, “science fiction drama”, respectively. As a result of comparative analysis, the proposed methodology outperformed the collaborative filtering-based recommendation system. From the results, we found that the time just prior to the recommendation was very important for inferring changes in user preference. Therefore, the proposed methodology not only can secure the transparency of the recommender system but also can reflect the users preferences that change over time. However, the proposed methodology has some limitations. The proposed methodology cannot recommend product elaborately if the number of products included in the topic is large. In addition, the number of sequential patterns is small because the number of topics is too small. Therefore, future research needs to consider these limitations.

종합 평점과 다기준 평점을 선택적으로 활용하는 협업필터링 기반 하이브리드 추천 시스템

구민정(Min Jung Ku),안현철(Hyunchul Ahn) 한국지능정보시스템학회 2018 지능정보연구 Vol.24 No.2

Recommender system recommends the items expected to be purchased by a customer in the future according to his or her previous purchase behaviors. It has been served as a tool for realizing one-to-one personalization for an e-commerce service company. Traditional recommender systems, especially the recommender systems based on collaborative filtering (CF), which is the most popular recommendation algorithm in both academy and industry, are designed to generate the items list for recommendation by using ‘overall rating’ – a single criterion. However, it has critical limitations in understanding the customers’ preferences in detail. Recently, to mitigate these limitations, some leading e-commerce companies have begun to get feedback from their customers in a form of ‘multicritera ratings’. Multicriteria ratings enable the companies to understand their customers’ preferences from the multidimensional viewpoints. Moreover, it is easy to handle and analyze the multidimensional ratings because they are quantitative. But, the recommendation using multicritera ratings also has limitation that it may omit detail information on a user’s preference because it only considers three-to-five predetermined criteria in most cases. Under this background, this study proposes a novel hybrid recommendation system, which selectively uses the results from ‘traditional CF’ and ‘CF using multicriteria ratings’. Our proposed system is based on the premise that some people have holistic preference scheme, whereas others have composite preference scheme. Thus, our system is designed to use traditional CF using overall rating for the users with holistic preference, and to use CF using multicriteria ratings for the users with composite preference. To validate the usefulness of the proposed system, we applied it to a real-world dataset regarding the recommendation for POI (point-of-interests). Providing personalized POI recommendation is getting more attentions as the popularity of the location-based services such as Yelp and Foursquare increases. The dataset was collected from university students via a Web-based online survey system. Using the survey system, we collected the overall ratings as well as the ratings for each criterion for 48 POIs that are located near K university in Seoul, South Korea. The criteria include ‘food or taste’, ‘price’ and ‘service or mood’. As a result, we obtain 2,878 valid ratings from 112 users. Among 48 items, 38 items (80%) are used as training dataset, and the remaining 10 items (20%) are used as validation dataset. To examine the effectiveness of the proposed system (i.e. hybrid selective model), we compared its performance to the performances of two comparison models – the traditional CF and the CF with multicriteria ratings. The performances of recommender systems were evaluated by using two metrics - average MAE(mean absolute error) and precision-in-top-N. Precision-in-top-N represents the percentage of truly high overall ratings among those that the model predicted would be the N most relevant items for each user. The experimental system was developed using Microsoft Visual Basic for Applications (VBA). The experimental results showed that our proposed system (avg. MAE = 0.584) outperformed traditional CF (avg. MAE = 0.591) as well as multicriteria CF (avg. AVE = 0.608). We also found that multicriteria CF showed worse performance compared to traditional CF in our data set, which is contradictory to the results in the most previous studies. This result supports the premise of our study that people have two different types of preference schemes – holistic and composite. Besides MAE, the proposed system outperformed all the comparison models in precision-in-top-3, precision-in-top-5, and precision-in-top-7. The results from the paired samples t-test presented that our proposed system outperformed traditional CF with 10% statistical significance level, and multicriteria CF with 1% statistical significance l

Evaluating the Quality of Recommendation System by Using Serendipity Measure

Tserendulam Dorjmaa,신택수 한국지능정보시스템학회 2019 지능정보연구 Vol.25 No.4

Recently, various approaches to recommendation systems have been studied in terms of the quality of recommendation system. A recommender system basically aims to provide personalized recommendations to users for specific items. Most of these systems always recommend the most relevant items of users or items. Traditionally, the evaluation of recommender system quality has focused on the various predictive accuracy metrics of these. However, recommender system must be not only accurate but also useful to users. User satisfaction with recommender systems as an evaluation criterion of recommender system is related not only to how accurately the system recommends but also to how much it supports the user’s decision making. In particular, highly serendipitous recommendation would help a user to find a surprising and interesting item. Serendipity in this study is defined as a measure of the extent to which the recommended items are both attractive and surprising to the users. Therefore, this paper proposes an application of serendipity measure to recommender systems to evaluate the performance of recommender systems in terms of recommendation system quality. In this study we define relevant or attractive unexpectedness as serendipity measure for assessing recommendation systems. That is, serendipity measure is evaluated as the measure indicating how the recommender system can find unexpected and useful items for users. Our experimental results show that highly serendipitous recommendation such as item-based collaborative filtering method has better performance than the other recommendations, i.e. user-based collaborative filtering method in terms of recommendation system quality.

사용자 리뷰 마이닝을 결합한 협업 필터링 시스템: 스마트폰 앱 추천에의 응용

전병국,안현철 한국지능정보시스템학회 2015 지능정보연구 Vol.21 No.2

Collaborative filtering(CF) algorithm has been popularly used for recommender systems in both academic and practical applications. A general CF system compares users based on how similar they are, and creates recommendation results with the items favored by other people with similar tastes. Thus, it is very important for CF to measure the similarities between users because the recommendation quality depends on it. In most cases, users' explicit numeric ratings of items(i.e. quantitative information) have only been used to calculate the similarities between users in CF. However, several studies indicated that qualitative information such as user's reviews on the items may contribute to measure these similarities more accurately. Considering that a lot of people are likely to share their honest opinion on the items they purchased recently due to the advent of the Web 2.0, user's reviews can be regarded as the informative source for identifying user's preference with accuracy. Under this background, this study proposes a new hybrid recommender system that combines with users' review mining. Our proposed system is based on conventional memory-based CF, but it is designed to use both user's numeric ratings and his/her text reviews on the items when calculating similarities between users. In specific, our system creates not only user-item rating matrix, but also user-item review term matrix. Then, it calculates rating similarity and review similarity from each matrix, and calculates the final user-to-user similarity based on these two similarities(i.e. rating and review similarities). As the methods for calculating review similarity between users, we proposed two alternatives - one is to use the frequency of the commonly used terms, and the other one is to use the sum of the importance weights of the commonly used terms in users' review. In the case of the importance weights of terms, we proposed the use of average TF-IDF(Term Frequency - Inverse Document Frequency) weights. To validate the applicability of the proposed system, we applied it to the implementation of a recommender system for smartphone applications (hereafter, app). At present, over a million apps are offered in each app stores operated by Google and Apple. Due to this information overload, users have difficulty in selecting proper apps that they really want. Furthermore, app store operators like Google and Apple have cumulated huge amount of users' reviews on apps until now. Thus, we chose smartphone app stores as the application domain of our system. In order to collect the experimental data set, we built and operated a Web-based data collection system for about two weeks. As a result, we could obtain 1,246 valid responses(ratings and reviews) from 78 users. The experimental system was implemented using Microsoft Visual Basic for Applications(VBA) and SAS Text Miner. And, to avoid distortion due to human intervention, we did not adopt any refining works by human during the user's review mining process. To examine the effectiveness of the proposed system, we compared its performance to the performance of conventional CF system. The performances of recommender systems were evaluated by using average MAE(mean absolute error). The experimental results showed that our proposed system(MAE = 0.7867 ~ 0.7881) slightly outperformed a conventional CF system(MAE = 0.7939). Also, they showed that the calculation of review similarity between users based on the TF-IDF weights(MAE = 0.7867) leaded to better recommendation accuracy than the calculation based on the frequency of the commonly used terms in reviews(MAE = 0.7881). The results from paired samples t-test presented that our proposed system with review similarity calculation using the frequency of the commonly used terms outperformed conventional CF system with 10% statistical significance level. Our study sheds a light on the application of users' review information for facilitating electronic commerce by recommen... 협업 필터링은 학계나 산업계에서 우수한 성능으로 인해 많이 사용되는 추천기법이지만, 정량적 정보인 사용자들의 평가점수에만 국한하여 추천결과를 생성하므로 간혹 정확도가 떨어지는 문제가 발생한다. 이에 새로운 정보를 추가로 고려하여, 협업 필터링의 성능을 개선하려는 연구들이 지금까지 다양하게 시도되어 왔다. 본 연구는 최근 Web 2.0 시대의 도래로 인해 사용자들이 구입한 상품에 대한 솔직한 의견을 인터넷 상에 자유롭게 표현한다는 점에 착안하여, 사용자가 직접 작성한 리뷰를 참고하여 협업 필터링의 성능을 개선하는 새로운 추천 알고리즘을 제안하고, 이를 스마트폰 앱 추천 시스템에 적용하였다. 정성 정보인 사용자 리뷰를 정량화하기 위해 본 연구에서는 텍스트 마이닝을 활용하였다. 구체적으로 본 연구의 추천시스템은 사용자간 유사도를 산출할 때, 사용자 리뷰의 유사도를 추가로 반영하여 보다 정밀하게 사용자간 유사도를 산출할 수 있도록 하였다. 이 때, 사용자 리뷰의 유사도를 산출하는 접근법으로 중복 사용된 색인어의 빈도로 산출하는 방안과 TF-IDF(Term Frequency - Inverse Document Frequency) 가중치 합으로 산출하는 2가지 방안을 제시한 뒤 그 성능을 비교해 보았다. 실험결과, 제안 알고리즘을 통한 추천, 즉 사용자 리뷰의 유사도를 추가로 반영하는 알고리즘이 평점만을 고려하는 전통적인 협업 필터링과 비교해 더 우수한 예측정확도를 나타냄을 확인할 수 있었다. 아울러, 중복 사용 단어의 TF-IDF 가중치의 합을 고려했을 때, 단순히 중

사용자 간 신뢰관계 네트워크 분석을 활용한 협업 필터링 알고리즘의 예측 정확도 개선

최슬비(Seulbi Choi),곽기영(Kee-Young Kwahk),안현철(Hyunchul Ahn) 한국지능정보시스템학회 2016 지능정보연구 Vol.22 No.3

Among the techniques for recommendation, collaborative filtering (CF) is commonly recognized to be the most effective for implementing recommender systems. Until now, CF has been popularly studied and adopted in both academic and real-world applications. The basic idea of CF is to create recommendation results by finding correlations between users of a recommendation system. CF system compares users based on how similar they are, and recommend products to users by using other like-minded people’s results of evaluation for each product. Thus, it is very important to compute evaluation similarities among users in CF because the recommendation quality depends on it. Typical CF uses user’s explicit numeric ratings of items (i.e. quantitative information) when computing the similarities among users in CF. In other words, user’s numeric ratings have been a sole source of user preference information in traditional CF. However, user ratings are unable to fully reflect user’s actual preferences from time to time. According to several studies, users may more actively accommodate recommendation of reliable others when purchasing goods. Thus, trust relationship can be regarded as the informative source for identifying user’s preference with accuracy. Under this background, we propose a new hybrid recommender system that fuses CF and social network analysis (SNA). The proposed system adopts the recommendation algorithm that additionally reflect the result analyzed by SNA. In detail, our proposed system is based on conventional memory-based CF, but it is designed to use both user’s numeric ratings and trust relationship information between users when calculating user similarities. For this, our system creates and uses not only user-item rating matrix, but also user-to-user trust network. As the methods for calculating user similarity between users, we proposed two alternatives – one is algorithm calculating the degree of similarity between users by utilizing in-degree and out-degree centrality, which are the indices representing the central location in the social network. We named these approaches as ‘Trust CF – All’ and ‘Trust CF – Conditional’. The other alternative is the algorithm reflecting a neighbor’s score higher when a target user trusts the neighbor directly or indirectly. The direct or indirect trust relationship can be identified by searching trust network of users. In this study, we call this approach ‘Trust CF – Search’. To validate the applicability of the proposed system, we used experimental data provided by LibRec that crawled from the entire FilmTrust website. It consists of ratings of movies and trust relationship network indicating who to trust between users. The experimental system was implemented using Microsoft Visual Basic for Applications (VBA) and UCINET 6. To examine the effectiveness of the proposed system, we compared the performance of our proposed method with one of conventional CF system. The performances of recommender system were evaluated by using average MAE (mean absolute error). The analysis results confirmed that in case of applying without conditions the in-degree centrality index of trusted network of users(i.e. Trust CF - All), the accuracy (MAE = 0.565134) was lower than conventional CF (MAE = 0.564966). And, in case of applying the in-degree centrality index only to the users with the out-degree centrality above a certain threshold value(i.e. Trust CF - Conditional), the proposed system improved the accuracy a little (MAE = 0.564909) compared to traditional CF. However, the algorithm searching based on the trusted network of users (i.e. Trust CF – Search) was found to show the best performance (MAE = 0.564846). And the result from paired samples t-test presented that Trust CF - Search outperformed conventional CF with 10% statistical significance level. Our study sheds a light on the application of user’s trust relationship network information for facilitating electronic commerce by rec

전자 저널 구독 정보 및 웹 이용 로그를 활용한 참고문헌 기반 저널 추천 기법

이해성 ( Hae-sung Lee ),김순영 ( Soon-young Kim ),김재훈 ( Jay-hoon Kim ),김정환 ( Jeong-hwan Kim ) 한국인터넷정보학회 2016 인터넷정보학회논문지 Vol.17 No.5

전자 학술 정보 유통의 확대에 따라 날로 증가되는 학술 콘텐츠 서비스 수요에 부응하기 위하여 보다 효과적인 학술 콘텐츠 추천 시스템 개발이 요구된다. 학술 콘텐츠 추천 시스템은 정보 소비자의 과거 이용 내역을 기반으로 각 소비자 선호(preference)에 맞는 학술 콘텐츠를 제공함으로써 콘텐츠 이용성을 보다 효과적으로 향상 시킬 수 있다. 본 논문에서는 특정 기관에 소속된 사용자의 선호에 더욱 부합하는 학술 콘텐츠를 제공하기 위하여 기관의 전자 저널 구독 정보 및 웹 이용 로그를 활용한 저널 추천 기법을 제안한다. 제안하는 추천 기법에서는 기관 사용자의 저널 선호도를 효과적으로 예측하기 위하여 기관 유사도(Institution similarity), 그리고 참고문헌의 인용 관계 데이터를 기반으로 저널 유사도(Journal similarity) 및 저널 중요도(Journal importance)를 산출하여 최종적으로 기관 맞춤형 저널 추천 항목을 구성하게 된다. 또한, 제안하는 추천기법이 적용된 기관 맞춤형 저널 추천 시스템 프로토타입을 개발한다. 개발된 저널 추천 시스템은 각 기관의 저널 선호도 예측을 위하여 활용되는 웹 이용로그를 효과적으로 수집하고 이를 추천 기법에 활용하기 용이한 데이터로 가공 처리 하여 별도의 데이터베이스에 저장하여 추천 기법의 저널 선호도 예측을 위한 기반 데이터로 활용한다. 마지막으로 우리는 기존 추천 기법들과의 비교 성능 평가를 통해 제안 기법의 차별성과 우수성을 보인다. With the exploration of digital academic information, it is certainly required to develop more effective academic contents recommender system in order to accommodate increasing needs for accessing more personalized academic contents. Considering historical usage data, the academic content recommender system recommends personalized academic contents which corresponds with each user`s preference. So, the academic content recommender system effectively increases not only the accessibility but also usability of digital academic contents. In this paper, we propose the new journal recommendation technique based on information of journal subscription and web usage logs in order to properly recommend more personalized academic contents. Our proposed recommendation method predicts user`s preference with the institution similarity, the journal similarity and journal importance based on citation relationship data of references and finally compose institute-oriented recommendations. Also, we develop a recommender system prototype. Our developed recommender system efficiently collects usage logs from distributed web sites and processes collected data which are proper to be used in proposed recommender technique. We conduct compare performance analysis between existing recommender techniques. Through the performance analysis, we know that our proposed technique is superior to existing recommender methods.

개인화 서비스 진전에 따른 자동추천 시스템 연구 동향과 방법론적 특성 연구

황용석(Yong-suk, Hwang),김기태(Ki-Tae, Kim) 사이버커뮤니케이션학회 2019 사이버 커뮤니케이션 학보 Vol.36 No.2

본 연구의 목적은 사회과학 연구자들을 위한 추천 시스템 연구의 연구 경향의 변화와 방법론적 동향을 탐색하는 것이다. 각각의 연구 경향에서 사용된 연구방법상의 문제점과 한계를 검토하고 추천 시스템 연구 범위의 확장과 더불어 새롭게 사용되는 연구방법들을 제시한다. 전통적인 추천 시스템 연구 경향은 크게 시스템 중심 접근법과 이용자 중심 접근법으로 나눌 수 있다. 그런데 시스템 중심 접근법의 주요 관심사여왔던 추천서비스 알고리즘의 예측 정확도 향상이 반드시 실제 이용자의 사용 만족도로 이어지지 않는다는 비판이 지속적으로 제기되었다. 이러한 이유에서 등장한 이용자 중심 접근법은 주로 전통적인 사회과학 실험 방법(lab experiments)을 사용하는데, 이러한 연구방법 역시 제한된 수의 피실험자들에게 제한된 변수만을 테스트하는 방식으로 도출된 결과의 낮은 신뢰도와 닞은 외적 타당성이 단점으로 지적되어왔다. 이 글에서는 현재 주요 글로벌 미디어 기업들(구글, 아마존, 페이스북 등)이 적극적으로 활용하고 있는 대규모 온라인 통제실험을 이러한 이용자 중심 접근법의 실험 방법론이 가지고 있는 약점을 극복할 수 있는 대안적 방법론으로 제시한다. 한편 추천서비스가 점점 우리 일상의 한 부분이 되어가면서, 기존의 추천 시스템 평가 위주의 연구영역에서 벗어나 추천 시스템의 사회적 영향에 대한 관심으로 연구 영역이 확장되었다. 이러한 연구 영역의 확장과 더불어 새로운 방법론적 시도가 이루어졌는데, 이 글에서는 온라인 뉴스 추천서비스 사용자들의 파편화 문제를 관계망 분석을 통해 다룬 일련의 연구들과 추천 알고리즘의 편향과 차별 문제에 데이터마이닝 기법과 대규모 온라인 통제실험방법을 접목한 연구들과 연구에 사용된 몇 가지 기법을 소개한다. The main purpose of this paper is to explore the recent trends of recommendation system research and methodological changes. This paper examined the issues and limitations of the research methods that each study in the filed employed. At the same time, this study suggested the newly emerging methodological trends, as the research area of recommendation system studies expands. Traditional recommendation studies are categorized into two approaches: system-centric approaches and user-centric approaches. However, recommendation system researchers suggested that the improvement of the prediction accuracy does not necessarily lead to user satisfaction of the system, which resulted in the changes in the approaches to the recommendation system studies from system-centric approaches to user-centric approaches. User centric approaches tend to involve traditional user experiments, with the low level of consistency across studies and external validity of the research design. Consequently, this paper suggested, as an alternative method, online controlled experiments at large scale which have been widely used by the major global media cooperation, such as Google, Amazon, Facebook. As recommendation systems has been integrated into our daily life, the area of interest in the field expanded to the issue of social influence of recommendation systems beyond the issue of evaluation of recommendation system, such as the issues of the fragmentation of the user groups of online news recommendation services and the bias and discrimination of the recommendation algorithm Together with such expansion of the research scope, new methodological attempt has been made. This paper examined how the newly emerging issue in recommendation research field are aligned with new research or analytic methods such as network analysis and data mining techniques.

협력필터링과 사회연결망을 이용한 신규고객 추천방법에 대한 연구

신창훈(Chang-Hoon Shin),이지원(Ji-Won Lee),양한나(Han-Na Yang),최일영(Il Young Choi) 한국지능정보시스템학회 2012 지능정보연구 Vol.18 No.4

Consumer consumption patterns are shifting rapidly as buyers migrate from offline markets to e-commerce routes, such as shopping channels on TV and internet shopping malls. In the offline markets consumers go shopping, see the shopping items, and choose from them. Recently consumers tend towards buying at shopping sites free from time and place. However, as e-commerce markets continue to expand, customers are complaining that it is becoming a bigger hassle to shop online. In the online shopping, shoppers have very limited information on the products. The delivered products can be different from what they have wanted. This case results to purchase cancellation. Because these things happen frequently, they are likely to refer to the consumer reviews and companies should be concerned about consumer’s voice. E-commerce is a very important marketing tool for suppliers. It can recommend products to customers and connect them directly with suppliers with just a click of a button. The recommender system is being studied in various ways. Some of the more prominent ones include recommendation based on best-seller and demographics, contents filtering, and collaborative filtering. However, these systems all share two weaknesses : they cannot recommend products to consumers on a personal level, and they cannot recommend products to new consumers with no buying history. To fix these problems, we can use the information which has been collected from the questionnaires about their demographics and preference ratings. But, consumers feel these questionnaires are a burden and are unlikely to provide correct information. This study investigates combining collaborative filtering with the centrality of social network analysis. This centrality measure provides the information to infer the preference of new consumers from the shopping history of existing and previous ones. While the past researches had focused on the existing consumers with similar shopping patterns, this study tried to improve the accuracy of recommendation with all shopping information, which included not only similar shopping patterns but also dissimilar ones. Data used in this study, Movie Lens’ data, was made by Group Lens research Project Team at University of Minnesota to recommend movies with a collaborative filtering technique. This data was built from the questionnaires of 943 respondents which gave the information on the preference ratings on 1,684 movies. Total data of 100,000 was organized by time, with initial data of 50,000 being existing customers and the latter 50,000 being new customers. The proposed recommender system consists of three systems : [+] group recommender system, [-] group recommender system, and integrated recommender system. [+] group recommender system looks at customers with similar buying patterns as ‘neighbors’, whereas [-] group recommender system looks at customers with opposite buying patterns as ‘contraries’. Integrated recommender system uses both of the aforementioned recommender systems to recommend movies that both recommender systems pick. The study of three systems allows us to find the most suitable recommender system that will optimize accuracy and customer satisfaction. Our analysis showed that integrated recommender system is the best solution among the three systems studied, followed by [-] group recommended system and [+] group recommender system. This result conforms to the intuition that the accuracy of recommendation can be improved using all the relevant information. We provided contour maps and graphs to easily compare the accuracy of each recommender system. Although we saw improvement on accuracy with the integrated recommender system, we must remember that this research is based on static data with no live customers. In other words, consumers did not see the movies actually recommended from the system. Also, this recommendation system may not work well with products other than movies. Thu